Monday, August 11, 2014

Setback for the No campaign as new YouGov poll fails to back up their complacent boasts of a post-debate polling boost

It's been a night of frustration for Blair McDougall, the No camp's embarrassment of a campaign chief.  He must have hoped that the new YouGov poll would replicate the post-debate increase in the No lead reported by Survation a couple of days ago, but instead it has shown no change whatsoever since the last poll from the firm several weeks ago.  Bear in mind that YouGov have in recent times overtaken Ipsos-Mori to become the outright most No-friendly pollster, largely due to the notorious "Kellner Correction" which artificially lowers the Yes vote, so the following figures should be seen in that light.

Should Scotland be an independent country?

Yes 39% (n/c)
No 61% (n/c)

On the figures that take account of Don't Knows there has been a 1% increase in the No vote, which is of no statistical significance at all -

Yes 35% (n/c)
No 55% (+1)

Now, to be fair, there are two ways in which the No campaign could argue that this poll is theoretically consistent with a substantial post-debate bounce for No, even if that's not showing up on the headline figures - but both are problematical for them.  The first way is to point out that, unlike the Survation poll, only half of the fieldwork actually took place after the debate, so it's possible that the post-debate part of the sample was much better for No.  The problem is that this would imply there must have been a swing to Yes in the pre-debate part of the sample, otherwise the overall Yes vote would be lower.  That's surely not something that McDougall Central would be comfortable with, because we know that debate bounces can recede very quickly.

The second way is to point out that the last two YouGov polls were probably understating the Yes vote, because they both showed swings to No at a time when other pollsters were detecting no such thing.  So it could be suggested that the Yes vote has actually fallen from 41-42% (with Don't Knows excluded), which is the high watermark YouGov showed in the spring.  But that would also be a problematical claim, because the No campaign were all too eager to insist that the changes shown in the last two polls were real, and they'd have to implicitly accept they were leading us up the garden path.

It's important to stress that YouGov are one of only two BPC pollsters (the other is TNS) who don't filter or weight their headline numbers by likelihood to vote.  However, like TNS, they do make additional turnout-adjusted figures available.  As far as I know, those figures haven't been released yet for this poll, so that'll be the first thing to look out for when the datasets appear, presumably tomorrow.  It's impossible to say whether the filtered results will be any different -  they weren't in the last YouGov poll, but in the last-but-one poll the No lead was a full 4% lower among respondents who say they will definitely vote.

It's been well-rehearsed that YouGov have acted in a thoroughly reprehensible manner since this long referendum campaign began.  When I or others raise question marks about the methodologies of various pollsters, it's sometimes asked : "Are you seriously saying that the pollsters are consciously biased?"  And generally the answer to that question is "no", but YouGov is a partial exception. I'd say they now occupy a no-man's-land somewhere between neutrality and outright pro-No bias.  Both Peter Kellner and Laurence Janta-Lipinski have published commentary on the referendum that has utilised familiar No campaign attack lines, and it's very hard to trust a firm capable of doing that to be scrupulously neutral in devising its methodology.

Nevertheless, just for once I can give a very small piece of credit to YouGov, because in this poll they have made two very sensible methodological changes, albeit arguably of the "too little, too late" variety.  They've followed in Panelbase's footsteps by introducing weighting by country of birth, which will have the effect of slightly reducing the No lead.  In this case it has only reduced what would otherwise have been a 62/38 split to 61/39, but once the hoo-ha over the debate fades away or goes into reverse, we may soon be back into a situation where a 2% or 4% decrease in the No lead will look a hell of a lot more significant.  Incidentally, if it's gradually becoming the new orthodoxy that weighting by country of birth is a wise idea, then it's highly likely that the No lead in the Survation poll should have been a bit lower.

YouGov's other change is to start interviewing 16 and 17 year olds, which is a ridiculously long-overdue step, because other pollsters have been doing it throughout the campaign.  It's very hard to understand how Kellner can try to pull rank on other pollsters when his own firm has been guilty of such an amateurish omission for so long.

Welcome though these changes are, neither of them can even begin to make up for the fact that the "Kellner Correction" remains in place, and the obsessive secrecy over the effect it is having continues.  Anthony Wells has a semi-detached relationship with YouGov that I don't fully understand, but his inside knowledge was sufficient to let us know instantly what the figures in this poll would have been if the new methodological changes hadn't been made.  And yet still no-one will tell us what the numbers would be if the "Kellner Correction" was removed.  When I asked Janta-Lipinski a few weeks ago, he initially tried to change the subject, before boasting about the firm's secretiveness and the fact that we weren't entitled to that kind of information.

More details on this poll, and a Poll of Polls update, can be found HERE.

58 comments:

  1. The largest lead for No since December, immediately after two events that were originally anticipated to produce a swing to Yes (the Commonwealth Games and the first debate). If that's a setback for No I'd hate to see a victory.

    ReplyDelete
  2. "I'd hate to see a victory."

    I don't blame you - Tory rule from Westminster is no joke.

    ReplyDelete
  3. An unchanged lead for no after the biggest propaganda offensive launched by the english media since WWII.

    Do you get london weighting for your spambottery or is it part of your internship?

    ReplyDelete
  4. Even the PB herd can see the truth now and again.

    "this poll reveals diddly squat and therefore is a waste of time, money and effort."

    Unlike the dismal haters who claimed that this shows no on the way to 60% of the vote and the end of democracy in Scotland forever.

    ReplyDelete
  5. "Do you get london weighting for your spambottery or is it part of your internship?"

    Heh heh.

    ReplyDelete
  6. "I don't blame you - Tory rule from Westminster is no joke."

    It's ok, I'm sure you'll be able to explain on 19 September why No getting more than 50% of the vote really means Yes won. The official results will have some fatal flaw in the rounding no doubt.

    ReplyDelete
  7. Anon : I see you've avoided the point about Tory rule. Does that mean you're a Tory yourself, or merely that you're embarassed to be an enabler for Tory rule?

    ReplyDelete
  8. What value is there is analysing polls when you're just going to spin everything in favour of Yes? I can't recall the last time when you haven't dismissed a No-friendly poll for being flawed or spun it as actually really good for Yes whilst any swing to Yes is treated as good news and gospel.

    I know this is a Yes blog but the shameless nature of it surely alienates all but the most myopic Yes supporters. You can want Yes to win but look at the poll and say "we need to improve the campaign" rather than "they need to improve the polls".

    ReplyDelete
  9. Kinda renders Kellner's blog on how Yougov were bang on the money a pile of dung.

    If you are right, then you don't need to change methods. If you believe you are wrong, you do. Of course you can't know whether the new changes make things any better until you get the final result at least.

    The CoB change is welcome but there's still a problem with it; a likely big one.

    If Yougov are still getting the very large disparity of too many English born, then up-weighting the Scots born who do answer is unlikely to correct the problem. They should be asking why not enough Scots born are responding, not giving the views of some respondents to a large group who are not responding / willing to be polled.

    This is a problem for all the online pollsters - why are scots not answering polls? Even MORI has something of a problem here, less as its telephone, but still there and manifesting as not enough Scots nat-ID'ing choosing to answer questions in addition to too many English born.

    What's causing Scots not to respond? One can only suggest shyness and most likely Yes based on the evidence we have for it from those who do respond.

    I believe it's only MORI now who haven't made some sort of big change. So, basically, cross-pollster agreement that they're not getting it right.

    ReplyDelete
  10. A troll using the word 'shameless' is well, pretty shameless.

    ReplyDelete
  11. I'm mentioned this before and possibly related to the under representation of Scots problem.

    In Yougov UK wide-polls, around the time of currencygate, the unweighted base for Scotland jumped from an average of 164 to 223. A lot more people were suddenly responding. This isn't a feature for the rUK regions.

    We can't know if the same is happening for Scotland-wide polls but it's likely.

    Odd that more are responding, yet less Scots are it seems?

    Might explain the CoB disparity anyway.

    ReplyDelete
  12. Wow, your analysis is amazingly one-eyed. Even allowing for your contention that YouGov is no-friendly, you're essentially trumpeting this "not as bad as we first thought" as a massive win.

    You're welcome to your pro-Yes view, but if you're really trying to be a serious commenter "one of Scotland's top 10 political websites" then you might try a bit harder to be rational.

    ReplyDelete
  13. As I mentioned on a past thread, if MORI is anything to go by, then the number of Scots born and/or Scots natIDing responding to polls has been decreasing over the past 2 years. However, this has not prevented Yes rising markedly across polls since the 2012-13 No peak.

    Samples becoming more 'British' yet more supportive of independence is a little odd.

    The simplest explanation is that Scots (more pro-Yes) are increasingly not responding, but those who are - and are more British - are moving to Yes.

    Certainly, a large group of the population deliberately not responding is a serious pollster issue. This always happens of course but normally this is a group who don't vote so it matters less.

    In the referendum case we have a group who were responding, but are increasingly not doing so. That's a whole different kettle of fish with big potential implications for pollsters.

    ReplyDelete
  14. I've often wondered what the polls ahead of 2011 would have been like if the press hadn't started to turn against Labour in the final stages, with the subway incident not getting air time etc.

    I'd propose that the polls would have shown much lower SNP gains, yet the result would have been the same on the day, giving an even bigger disparity between these two.

    The SNP big win, set from 2009, was clearly seen in government satisfaction. Even though labour were hitting high 40's in VI (vs ~30% SNP) early 2011 and were apparently well ahead of the SNP, they were not rated and instead the SNP were on 50-60% satisfied throughout. It was this number that told you what was going to happen, not VI.

    The press partly turning against Labour at the end just allowed people to justify why they were going to vote SNP and so admit it.

    Of course they had been admitting it to SNP canvassers on doorsteps already; hence the disparity between canvassing and polls.

    If such a pattern is repeating here, given the much greater sensitivity of the issue, we would expect the effect to be even more exaggerated. This may be why Yes believe the result may not be indicated in polls until very very close to the day.

    It was t-23 in 2011 before the SNP took a large lead and even then this lead looked shaky as the gap narrowed in the next poll from 10 points to 3. Only by t-11 did a clear SNP win look to be the outcome with no poll getting the scale of it correct.

    ReplyDelete
  15. When the polls blew it on the 2011 election, they lost some credibility and are now seen by many people as a tool of Westminster or the MSM whom they do not trust.

    If they call it wrong again, they will no doubt blame those weird Scots rather than themselves, but once it gets onto social media...

    I see that as well as YouGov, Ipsos Mori is now in hot water.

    Look at: http://munguinsrepublic.blogspot.co.uk/

    bjsalba










    ReplyDelete
  16. "When the polls blew it on the 2011 election"

    They didn't blow it per se; rather they just reported what people were saying they planned to do. A large section of the electorate were not honest however, and said Labour when they planned SNP. This section did say SNP to canvassers, just not to pollsters, not until the press started to give them a handy means to justify anyway.

    The giveaway in polls was when you asked these people how satisfied they were with the SNP/AS vs Labour/Gray. Here, they said they didn't rate Labour but rated the SNP. An odd thing to do unless you planned to vote SNP.

    It would be great if people didn't shy away from answering polls and everyone was totally honest. That's not the case however. This happens in all polls but is a particular problem on sensitive issues; people are human after all. You couldn't get more sensitive an issue than this one; even 2011 doesn't come close and look at the scale of porky-pie telling that went on for that.

    Maybe I'm wrong, but that would make Scotland weird as hell. It's if I'm right that it's behaving normally.

    ReplyDelete
  17. Bar the warriors, a normal, rational human reaction to being battered / verbally abused incessantly is to keep quiet / your head down until you can make your escape.

    This applies to a population was a whole as much as to an individual. Scotland is after all a person; just one with 5.3 million minds.

    Scotland has been getting screamed at since 2007. What did people expect?

    God knows what it would have been like if people had been been more honest and Yes had been ahead as per 2011 and 1997-07. They'd probably already have bulldozers out pretending to start work on border posts.

    ReplyDelete
  18. But much of the polling was done before the debate?!

    ReplyDelete
  19. A picture tells 1000 words. See picture which shows credibility of YouGov's polls in Scotland.

    https://twitter.com/JamesCo77979225/status/498762285230612480/photo/1

    ReplyDelete
  20. 10/10 certain to vote:

    40(+1)% Yes
    60(-1)% No

    Imagine that Scotland votes NO this September, do you think...

    39% The matter should be considered settled and their should not be another referendum
    53% There should be another referendum


    (25% within 10 years, 11% within 10-15, 17% within 20-30)

    That's not very good for No. A no friendly pollster yet a minority think a No should settle the matter with a majority thinking another referendum should be held, most saying within 15 years).

    And of course the poll finds Labour ahead of the SNP due to Kellner correction making Yougov very lonely in this respect and of course boosting No. Even MORI has the SNP clearly out in front.

    Details of the Kellner correction remain hidden. 5.2% to many English. How can they not get a proper sample unless Scots are not answering?

    ReplyDelete
  21. I'm voting YES but

    61---NO
    39---YES


    is still nae good

    ReplyDelete
  22. Re: Scottish_Skier 10.45 AM

    So that is at least 53% of respondents who, when they think about it, don't really want a No vote in September.

    ReplyDelete
  23. Scottish skier wrote

    "In 2011, the first poll to suggest the SNP would win clearly was at t-23 (we are t-38). The gap was then closed again between the SNP and labour at t-14. Only after t-11 did the polls start to show in earnest what would possibly happen"


    I don't recognise these figures for 2011 at all but I think they come from the regional vote which is irrelevant here as it involves a complicated pollimg formula involving PR and second choices The polling of the consittuency vote (First Past the Post as here) in 2011 is the one that is relevant for this referendum and it tells an entirely different story

    In the polling in 2011 the parties were leval or SNP slightly ahead from t-40. If there is to be an analogy that should be showing up now, it does not appear to be but I thought it was why we were watching every poll with eagle eyes. It's certainly why I am. No doubt you will say the media have to attack BT in order to allow shy Yes to come out but is there any polling basis for this or is it blind faith? Are there figures to demonstrate your claims people lied to pollsters? If so where are they because without them it's surely an opinion, in my view a fantastical one.

    In 2011 after t-40 the SNP went into a big polling lead as early as t -28 (Ie in 9 days time on this timescale) and four weeks before polling day. The idea the polls were miles off in the last month is not right.

    To find the SNP as far behind as they are now in a poll then in 2011 you'd have to go back to t-67. And as I say after t-28 SNP was way ahead for the rest of the contest in every poll bar just one that put them only three points ahead on t-21.

    So we are nearly at the stage now (in fact past it) when the polls should change dramatically if there is to be any similarity.

    And my own view (though I cannot prove it any more than I think you can prove people are lying in vast numbers) is that 2011 measured a very volatile election where people were voting for a few years of government. Here it's for ever and my guess is the core No vote is going to be pretty near as imnmovable as core Yes..

    ReplyDelete
  24. James: "I see you've avoided the point about Tory rule. Does that mean you're a Tory yourself, or merely that you're embarassed to be an enabler for Tory rule?"

    It stands to reason that if the equivalent of 61% of the electorate are voting No to independence, and there are only around 15-20% of people who are genuine Tories in Scotland, then we can't all be Tories.

    Yet for some unknown reason we keep trying to portray all No voters as Margaret Thatcher loving, Tory-voting, "britnats". Is it really that hard to accept people can look at the issue seriously and come to a legitimate difference of opinion without them all being Tories?

    ReplyDelete
  25. Anon : Nice try with the straw man : now answer the questions I actually asked. Are you a Tory? If not, are yoi embarrassed about being a Tory enabler?

    I take it from what you say that you do actually live in Scotland? Whereabouts?

    ReplyDelete
  26. Those us who were in scotland and campaigning in 2011 know perfectly well just how late and how little the pollsters eventually caught up to reality. It's why the canvassing and reports from the ground were bang on.

    Pretending that 2011 didn't show the pollsters, unionist media and pundits feet of clay for all to see is amusing, but no more than that.

    The continued methodology changes only weeks away from the poll and the preposterous "Kellner Correction" from YouGov hardly indicates anything other than a pollster who is very far from confident in it's own figures and methods.

    Nonetheless, if the No campaign really are convincing themselves that they are comfortably ahead then their already sketchy and limited plans for the crucial ground campaign and GOTV is going to be that much harder for them to maintain in the final vital weeks. That when they absolutely could not afford for their own limited number of activists to take the result for granted.

    Yes supporters and activists will obviously not let the backdrop of unionist media hysteria and some questionable polling stop us from turning out in force. It didn't in 2007, it didn't in 2011 and it certainly won't now for an Independence vote we have worked tirelessly for decades to make a reality.

    We are somewhat used to being told what we can't do by out of touch westminster politicians and their inept cheerleaders. We were always told we couldn't have this referendum, that it was a pipe dream, that it would never happen in our lifetimes.

    Yet westminster was proved wrong and we were the ones who were proved right.

    ReplyDelete
  27. James: "Nice try with the straw man : now answer the questions I actually asked. Are you a Tory? If not, are yoi embarrassed about being a Tory enabler?

    I take it from what you say that you do actually live in Scotland? Whereabouts?"

    This is honestly the best you can do? Badgering people about whether they're Tories or live in Scotland?

    No I'm not a Tory. Yes I live in Scotland. What exactly that is supposed to prove is beyond me. Why don't you try coming up with some kind of coherent argument about the referendum rather than ad hominem gibberish.

    ReplyDelete
  28. "Are there figures to demonstrate your claims people lied to pollsters?"

    That's impossible to prove. You can't ask people if they are lying to you - they'll say they are not. SNP were on 40% by 2009. All they needed was the coalition to get that extra 5%. Support for them never went away 2009-early 2011, as evidenced by satisfaction ratings (have a look at MORI)and canvassing. People said Labour when they planned SNP. Labour were already bottoming out by 2003. That massive surge 2009-early 2011 was a mirage. It was not coordinated, but all over Scotland a group of people did the same thing. I've suggested reasons for this.

    Yes was solidly ahead 1997-2007. Scotland then voted 40% for indy parties. It then voted >51% for pro-indy parties. Did it go off independence in that time whilst moving towards a referendum? That would be a little odd, particularly given Yes went ahead again in 2011.

    Some people must have been saying Yes for over 10 years to 2007, then said 'well No/DK' for a while to 2011, then Yes again in 2011, then No again 2012-13, then increasingly Yes again 13-now to explain the long term trend.

    Maybe they really do keep chopping and changing, but that's very odd as you say because this is not an election and most peoples leanings on the subject are something they've held for a long time.

    Are they chopping and changing or are they responding based on how acceptable it is to say Yes? I can't prove either.

    What I do no, is that if we had No Y/N polls (or at least none of the confusing ones from 2007 on), only ones on government trust (Holyrood and not Westminster big style!!!!!, who people want to run Scotland (Holyrood again!), how Scottish Scotland is vs British (62-74% rock solid Scottish), whether it would vote to join the UK today (big No), whether people think it would be a successful independent nation (big Yes)... added to the political history of constantly moving closer and closer to independence over the past 70 years... you'd conclude a Yes vote was most probable.

    Maybe people are being totally honest in Y/N polls. We no some are not (ICM people saying No then marking themselves as a Yes) it would seem, but lets forget them and assume complete honesty. That would make Scotland unique in the world; the only place where people don't lie to pollsters about even the most trivial things.

    Coming back to 2011. My records have t-35 before we had the start of the SNP consistently ahead or equal. Only at t-23 did we get our first big lead poll and that was when the real, big shift began. There was one MORI poll at t-77 which had the SNP fractionally ahead; this was dismissed based on falling SNP and rising Labour in following polls.

    We have already had Yes ahead, maybe mirroring this, just in polls where we offer people a justification for Yes in the question so they don't need to provide one themselves, e.g. Tories (or even Labour) back in 2015 and a possible Yes majority. No in some polls is less than Yes in others too; we have cross-over and have had that for some time, just not in a single straight Y/N.

    If Scotland was like e.g. Catalonia which controls its own press, I very much doubt we'd have any evidence for shy Yes and instead things might be much clearer from polls. Scotland's situation is quite unique in this respect. It is to all intents and purposes 'occupied' by the British state in terms of media control / bias. A heads down response would be normal in that sense globally.

    I cannot guarantee I am right, but I'm prepared to stick my neck out and offer evidence to support this as I have been doing.

    ReplyDelete
  29. ...Even today's Yougov is odd. Good for No due to methods yet a No doesn't put the matter to bed and a majority of people want another referendum? That doesn't make much sense unless people were rather more inclined to indy than they were letting on.

    It was the same when yougov asked about the Rowling effect; it boosted Yes when you'd think the opposite would be the case?

    Anyway, there's either truth to this and we're in for a surprise, or it'll be a tight race. We'll just have to wait and see. I personally see no reason for Yes to be despondent. If we get to t-7 and things have not moved at all, I'll start to be concerned.

    ReplyDelete
  30. Anon : Why are you finding these very reasonable questions so uncomfortable? If you do live in Scotland, whereabouts? If you're not a Tory, which party did you vote for in the 2011 election? Would you describe yourself as left-wing, right-wing or centrist?

    ReplyDelete
  31. Some thoughts:

    http://loveandgarbage.wordpress.com/2014/08/11/a-few-thoughts-on-referendum-opinion-polling/

    ReplyDelete
  32. See all you no trolls on here who scream "distortion, bias, desperation" are you actually claiming that James's analysis is wrong and that the polls are correct?

    Heh heh

    You should have a read of the analysis, then show us one single instance of where he has made a factual error. You can't can you.

    ReplyDelete
  33. Re Scottish skier

    We aren't exactly in disagreement, as you say it's an odd kimd of entity in polling terms, this referendum and your opinion of what is going on is as good as anyone's

    But I would still query your 2011 figures, The first poll alerting Scotland on what was to come with a big SNP lead was t-26 and as I say we are only about ten days from there now on that time-scale. After that the SNP stayed well ahead

    Meanwhile at this stage today ( t-37.) the last equivalent poll in 2011 had put the SNP ahead (admittedly by a measly one percent and it was YOuGov!) but there seemed everything and more to play for for the SNP. It was nothing like this is, the polls were absolutely and clearly on a dead heat.

    They just aren't right now though they may be next week.

    ReplyDelete
  34. Expat.. just to add a reminder that what I've been saying comes from the pro-union camp too.

    Here's Alistair Carmichael:

    “If you know you shouldn’t do something but for an emotional reason you are going to do it anyway, then you won’t admit to pollsters what you are going to do.

    “You’ll say you haven’t made your mind up, but know exactly what you are doing but won’t admit it because you are making a judgement that is emotional rather than intellectual.”


    http://www.dailyrecord.co.uk/news/politics/independence-referendum-could-go-way-2797015

    Now his given reasoning is coming from a unionists perspective so isn't quite right, but he's aware that No's lead isn't quite what it seems. I imagine he's got this from in-depth analyses of internal polling.

    ReplyDelete
  35. They just aren't right now though they may be next week.

    I would expect some movement to Yes in the final weeks if I'm at all correct. There might be none, but that would be unlikely if Yes was going to win.

    If e.g. a big tabloid came out for Yes such as the Sun, then bigger movements. Not because the Sun was changing minds, but rather making it more acceptable / mainstream to admit Yes.

    I'm not a great believer in the power of the [traditional] media to influence what people believe. the internet and social media has greatly weakened this. The UK for example has some of the lowest levels of media trust in Europe and polls show trust in papers is at an all time low in Scotland.

    The idea the SNP could have got to where they are if people took the papers as gospel just doesn't add up for example. However, the traditional media do still have the power to influence how acceptable people feel it is to publicly express a view they have on something.

    Demonisation of Yes is designed to stop people talking about independence and to suppress it in polls, whereby making it appear Yes people are a minority / alone. That hopefully will take the wind out of Yes sails. It is a desperate tactic and one which has proven ineffective on final results, e.g. 2007 + 2011.

    I'm confident we do have a group of shy Yes / Soft No; the circumstantial evidence is considerable, so much so even the unionists have admitted it exists.

    How big this group is and whether it's enough to carry a Yes is much harder to say. I believe it should be sufficient and if it exists at levels polls suggest, then a solid Yes win is possible... but I'm an eternal optimist ;-)


    ReplyDelete
  36. Blinking HystericalAugust 11, 2014 at 1:44 PM

    Scottish_Skier: 2007 and 2011. People lying to pollsters not pollsters lying to people? Why do you think the former is what happened and not the latter?

    ReplyDelete
  37. Smithson major has a post up claiming that the new kellner fix makes the rigged u-gov polls more favourable to Yes. Which doesn't seem at all likely given the facts.

    However I'm sure that the uk's foremost political bigot, I mean punter knows what he's talking about from his english nationalist libdem perspective.

    ReplyDelete
  38. Anon : The new methodology is indeed slightly more favourable for Yes - I pointed that out in the blogpost. In this case it has reduced what would otherwise have been a 62-38 No advantage to 61-39. In future polls it's conceivable that it will sometimes reduce the No lead by as much as 4% on the rounded numbers. Whether Smithson likes it or not, that's entirely justified, because YouGov have far too many English-born people in their panel, and without weighting those respondents down they would be producing skewed results (or even more skewed, perhaps I should say).

    ReplyDelete
  39. Expat, do you blog at all? I would be interested in reading it?

    ReplyDelete
  40. The Sun is mentioned here. Is it not the case the Sun supports YES for reasons of the most utterly suspect kind? Essentially Murdoch wants to get his own back on the UK. And my feeling is his tabloid it will certainly endorse YES near polling day if it has not already.

    And as NO I have to live with the support of the Mail, that's painful enough! You are welcome to Murdoch.

    But if he is gunning Yes (and I'm sure he will) it hardly seems you can still fly the idea all of the media is against you.


    ReplyDelete
  41. Expat : The Scottish Sun is neutral at the moment, to the best of my knowledge. The English Sun appears to be anti-independence (and its political editor is a rabid Nat-basher).

    ReplyDelete
  42. I notice that our Kellner correction group has changed from:

    SNP (but Labour in 2010)

    to:

    SNP (Holyrood) & Lab (Westminster).

    Does this mean our group has been changed from some sort of Labour 2010 tactical then SNP 2011 to a fully fledged 'We love Labour - we just vote SNP for Holyrood', as created by the unionists post 2011, so Scots can be expected to flock to Labour again in 2015 as Kellner wants them too, adjusting the poll samples to suit?

    ReplyDelete
  43. To be honest, I reckon there is a bit of truth in the kellner Correction, but only a bit.

    My own view is that it is half and half. I am an SNP voter but I voted labour in 2010 to keep the tories out.

    My brother in law is a labour voter, he voted labour in 2010 but SNP in 2011 to keep labour out (I know...) because of things like the poll tax freeze, NHS but mostly student fees. Mind you I'm not so sure how labour he is any more, he is definitely voting yes.

    How can anyone possibly claim to know what the truth is? Its all pure guesswork.

    GrahamMcG

    ReplyDelete
  44. Expat - for 2011 polls, why favour constituency over list?

    Constituency includes tactical voting and local factors, eg. personal vote for well regarded MSP. List vote is purely which party you support.

    Constituency lends itself to the shy voter as described by scottish_skier, they have a ready made "excuse" to vote for the "racist" party: "I'm doing it to keep out." No such excuse for the list vote.


    scottish_skier - why do you think a shy vote will reveal itself close to the day? Why not just keep head down until actually in the polling booth?

    ReplyDelete
  45. btw, the SNP list VI in 2011 polls only once showed over 40% (MORI at t-14) in the 2 years before the election. Actual result - 44%.

    There is still hope. ;-)

    ReplyDelete
  46. "scottish_skier - why do you think a shy vote will reveal itself close to the day? Why not just keep head down until actually in the polling booth?"

    Ah, now couldn't this be an indication that the pollsters were distorting things in 2011 rather than voters lying?

    The pollsters then start to report more accurately nearer the time so that it looks like they were right all along and it was the voters who were lying...

    ReplyDelete
  47. If Kellner thinks it is a "we love Labour, we just vote SNP for Holyrood" how does he explain that only a quarter of us voted for them in the Euro elections? Doesn't appear that we love Labour after all.

    ReplyDelete
  48. A little point to make. I see lots of Yes groups conducting voter registration drives, however on the No side this is being completely ignored? It's as if they know that this group will be voting heavily Yes.

    I'll bet my bottom dollar the polls are not picking up the views of these people.

    As Jim Sillars correctly states, this will be won in the housing schemes of Scotland not the chattering middle classes.

    ReplyDelete
  49. Kellner correction aside, how do you manage this?

    2011 voter Y/N support net (%Y-%N) from latest polls:

    SNP net Yes
    +59 Survation
    +56 TNS
    +51 ICM
    +50 Panelbase
    40 Yougov

    Labour net Yes
    -29 Panelbase
    -30 Survation
    -40 ICM
    -49 TNS
    -60 Yougov

    Lib net Yes
    -51 Survation
    -56 Panelbase
    -60 TNS
    -64 Yougov
    -70 ICM

    Tory net Yes
    -81 Survation
    -87 Panelbase
    -82 ICM
    -91 TNS
    -91 Yougov

    That can't be just random. The others show no clear order other than panelbase and Survation tending slightly more to get higher Yes results - only Yougov right down at the bottom each time.

    Yougov Yes by party lower even than TNS which has the lowest Yes of all the polls.

    If you hand pick your sample you could achieve that. I've heard lots of people say once they said they were Yes they weren't asked again by Yougov. Happened to me - I answered a few then nothing but soft drinks and telly. Maybe there is some truth to it.

    I'd have not considered a tinfoil hat until the Kellner correction, but now I'm actually open to the possibility. Unless there's an explanation for the above?


    ReplyDelete
  50. Reply to James who wrote

    "The Scottish Sun is neutral at the moment, to the best of my knowledge. The English Sun appears to be anti-independence (and its political editor is a rabid Nat-basher)".

    Hmmm interesting. I am highly sceptical of any Murdoch 'neutrality'. Also he likes to spring surprises and wield the greatest possible power and maximum shock effect. If the English editor does a bit of nat bashing that could only increase the surprise when it comes.

    My prediction would be Murdoch intends right now to have the Scottish Sun go from 'neutrality' to 'Yes' at the moment of greatest possible impact and maximum disadvantage for the No camp. That is his style (and he loathes Cameron and all the UK political parties who have taken him on at last). A week before polling maybe?. Or after the second debate.?

    There is however one big BUT. If Murdoch decides No will win, he may well abandon the plan. He never likes to back the losing side.

    (His support reinforces all my doubts about Yes though I accept no Yes supporter will care about that if he scores for them)

    . Brian said...
    Expat - for 2011 polls, why favour constituency over list?

    Constituency includes tactical voting and local factors

    Fair point, local factors of course but in the end they add up to a national vote.

    And I think that minor problem in the constituency polling is massively outweighed by the fact the regional vote is conducted by PR which gives a completely different kind of result, involving lists and multiple choices and is agonizingly difficult to poll. In fact I am kind of amazed it can be done at all and not remotely surprised when it is done badly. So extrapolating from that Regional vote to this vote is comparing eggs with oranges. In contrast the constituency vote like this vote should be fairly accurately poll-able and for the last month of the 2011 campaign (in my opinion) it largely was.







    ReplyDelete
  51. Scottish_Skier said...
    Kellner correction aside, how do you manage this?

    Sorry can you explain what this is, I don't understand these figures or what they reflect? You'd have to elaborate.

    ReplyDelete
  52. I might also add that since the creation of the Kellner correction, the need to up-weight this group has steadily decreased (and helped No as a result).

    'If you build it they will come'. Just like that. Magic.

    Or, Yougov have made this group up (true) and have gone out hunting for them because quasi-random sampling isn't producing the goods as the group isn't anywhere as big as Yougov hopes. Each time they find one, they record and return to the same people asking them again and again?

    The fact that they've spent the whole of 2014 so far trying to find these elusive Lab-SNP people and still not managed it would suggest they're very thin on the ground. Yet their views are being handed to 10% of the sample.

    ReplyDelete
  53. Expat: "And I think that minor problem in the constituency polling is massively outweighed by the fact the regional vote is conducted by PR which gives a completely different kind of result, involving lists and multiple choices and is agonizingly difficult to poll."

    Have you voted in a SP election?

    The list vote is just put one cross beside your party of choice. No multiple votes or preference or any of that gubbins. Simples.

    ReplyDelete
  54. Scottish Skier, you have appeared to say a couple of times that YouGov not finding enough Scottish born voters on their panel is something to do with Scots not responding.

    I am on the YouGov panel as are 2 other Yes voting family members. None of us have been asked to take part in a published political poll for YouGov for months. They already know we are confirmed Yessers and thus can, if they so wish, easily not invite us.

    Just seems an easy way for them to get the results they want. The stakes are so high that it seems naive to believe that pollsters can be assumed to be impartial.

    ReplyDelete
  55. I certainly wouldn't think someone who'd write this could be seen as impartial:

    http://www.prospectmagazine.co.uk/magazine/what-salmond-will-do-next-snp-referendum#.Uc_QF-BfXHQ

    ... but I tend to think the other firms are more interested in getting it correct than getting it "right", if you know what I mean?

    ReplyDelete
  56. Sorry Brian, re regional vote. apologies there are no preference lists but it is still complicated enough and devilishly tricky to poll as is shown by the following description.

    56 regional seats; these seats are awarded using a formula. The formula is the total number of regional votes received by a party or independent candidate divided by the number of seats (constituency and regional) already gained in that region +1. The party with the highest result after the formula is applied gain an additional seat. The calculation is repeated until all the additional seats have been awarded. So, for a party with no seats the number of votes received is divided by one, and so stays the same. If the party already has one seat in that region then its number of votes is divided by two, if it has two seats in that region it is divided by three, and so on. - See more at: http://www.aboutmyvote.co.uk/how_do_i_vote/voting_systems/scottish_parliamentary_electio.aspx#sthash.RnE5nwyt.dpuf. with

    ReplyDelete
  57. "Sorry can you explain what this is, I don't understand these figures or what they reflect?"

    They are net support for independence based on what people said they voted in 2011. The number is %Yes - %No amongst each group of voters (for the last poll from each).

    Curiously, for Yougov, for all groups, the level of support for Yes is the lowest or near so (lib dem samples are tiny), very notably for the SNP and Labour.

    If Yes was just lowest in SNP voters, we might think they have an issue with them. But it's all party supporters. Really, there should be some variation between pollsters, but general agreement. If e.g. panelbase got the highest Yes for every group we'd be suspicious too.

    That needs explaining. Are more no leaning voters all flocking to yougov? That makes no sense. The Yes scores for online, more anonymous Yougov are even lower than for face to face / chap your door TNS. Something very weird with Yougov.

    ReplyDelete
  58. Expat, you are describing the process for accumulating and measuring the votes once they are made not the actual vote itself.

    As Brian correctly points out the vote itself is simple enough and hardly requires an extensive knowledge of the process then used to tally it up for the result.

    ReplyDelete